Suggestion does not de-automatize word reading: Evidence from the semantically based Stroop task

Recent studies have shown that the suggestion for participants to construe words as meaningless symbols reduces, or even eliminates, standard Stroop interference in highly suggestible individuals (Raz, Fan, & Posner, 2005; Raz, Kirsch, Pollard, & Nitkin-Kaner, 2006). In these studies, the researchers consequently concluded that this suggestion de-automatizes word reading. The aim of the present study was to closely examine this claim. To this end, highly suggestible individuals completed both standard and semantically based Stroop tasks, either with or without a suggestion to construe the words as meaningless symbols (manipulated in both a between-participants [Exp. 1] and a within-participants [Exp. 2] design). By showing that suggestion substantially reduced standard Stroop interference, these two experiments replicated Raz et al.'s (2006) results. However, in both experiments we also found significant semantically based Stroop effects of similar magnitudes in all suggestion conditions. Taken together, these results indicate that the suggestion to construe words as meaningless symbols does not eliminate, or even reduce, semantic activation (assessed by the semantically based Stroop effect) in highly suggestible individuals, and that such an intervention most likely reduces nonsemantic task-relevant response competition related to the standard Stroop task. In sum, contrary to Raz et al.'s claim, suggestion does not de-automatize or prevent reading (as shown by a significant amount of semantic processing), but rather seems to influence response competition. These results also add to the growing body of evidence showing that semantic activation in the Stroop task is indeed automatic.